| In the dissertation, new methods for the determination of the trace arsenic and its speciation in environmental samples are developed using flow injection gas phase chemiluminescence.The dissertation is divided into three parts.In the first part, techniques, including chemiluminescence, gas-phase chemiluminescence and flow injection analysis are reviewed with the emphasis on their application on environment analysis.In the second part, a new method for the determination of trace arsenic in environmental samples is developed by flow gas phase chemiluminescence. Arsenic is reduced to arsine by sodium borohydride in 0.6mol/L sulfuric acid using flow injection analysis technique. Gaseous arsine is separated with a home-made gas-liquid separator and carried to the gas-phase reaction chamber, where it reacts with ozone to generate chemiluminescene. Chemiluminescene is detected by a photomultiplier tube. Under optimal conditions, a good and wide linear relationship (r=0.9996) is obtained in the range of 0.5 to500μg/L of arsenic with a sampling frequency of 50/h. The limit of detection is 0.10μg/L and the relative standard deviation is 0.77% (20μg/L As, n=11). The method is applied successfully to the analysis of arsenic in environmental samples and the results are in good agreement with those obtained by hydride generation atomic absorption spectrometry.In the third part, a method for the arsenic speciation in environmental samples is developed by gas phase chemiluminescence. It is based on the different pH dependence of the yield of arsine. Both As (III) and As (V) are quantitatively converted to arsine in the presence of 1.5% borohydride when 0.60 mol/L sulfuric acid is manually chosen, while only As (III) is converted to arsine when 2.0 mol/L citric acid-citrate (pH=4.8) is used. The generated arsine, driven by argon, is separated from solution in the gas-liquid separator and produces a gas-phase chemiluminescence by ozone in the reflective cell of the analyzer. Flow injection analysis technique is used for hydride generation which enables reproducible and controllable conditions. Under optimal conditions, the limit of detection is 0.33μg/L As (V)and 0.12μg/L As (III) respectively. The sample throughput was 50 /h. The method has been applied successfully to the analyses of As (III) and As (V) in environmental samples and the results are in good agreement with those obtained by hydride generation atomic absorption spectrometry. |